History of Excavations
The Danish Expedition (1957)
Tell Shemshara was first excavated in 1957 as part of the Dokan Dam salvage project. The Danish Dokan Expedition was organised by Jørgen Læssøe, with Harald Ingholt as field director. The team arrived at the site on 19 May 1957 and worked for ten weeks through the summer heat.
Excavations focused on the high northern part of Main Hill, where the team uncovered a sequence of sixteen occupation levels, ranging from the Neolithic (Hassuna-Samarra period) to the Medieval era. In mid-July, a small trench was opened on the lower southern extension of the mound. Here, just a few days before the end of the season, the team discovered the "Archive of Kuwari": approximately 150 cuneiform tablets dating to the early 18th century BCE, found in a room with a baked brick floor that had been destroyed by fire in antiquity.
Political unrest in Iraq prevented the Danish team from returning in 1958.
Tell Shemshara during excavation in summer 1957. View from north towards the high summit (photo taken by Friis, courtesy of the National Museum of Denmark).
Gallery: The Danish Expedition, 1957
Læssøe (centre) in the Danish camp at Shemshara (flanked by two unidentified visitors – probably engineers from the Dokan Dam construction site).
The Danish team in 1957 included an ethnographer, Henny Harald Hansen, who later published a highly interesting book The Kurdish Woman's Life. Hansen mostly stayed in villages outside the Rania Plain, but paid occasional visits to the team base.
Photo of a Kurdish woman cooking food on a fireplace, taken by Henny Harald Hansen during her research in the Rania Plain.
The room (No. 2) in the Palace where the Danish expedition found the archive of Kuwari (photo Friis, courtesy of the National Museum of Denmark).
One of the sherqatis (master workmen from the village of Sherqat) clean mudbricks at Shemshara (photo Friis, courtesy of the National Museum of Denmark).
The sherqatis working at Shemshara packing finds at the end of the season in 1957 (photo Friis, courtesy of the National Museum of Denmark).
The Iraqi Excavations (1958–1959)
Work at Shemshara continued under the direction of Abd al-Qadir al-Tekriti. During the 1958 season, Iraqi archaeologists exposed much of the Level V palace where the Danish team had found the archive. In two rooms south of the central courtyard, they discovered a second group of tablets—about 100 pieces, mainly administrative lists concerning agricultural products.
The team also conducted excavations on Camp Hill, where they uncovered remains from the Mitanni period (Late Bronze Age). In 1959, further work was carried out at Shemshara and at the nearby sites of Dugirdkan and Gird Bardastee. By 1960, Lake Dokan had formed and all archaeological activity in the area came to a halt.
Shemshara 1958: general overview of the Level V palace excavation, viewed from north.
The NINO Project (2012–2017)
Over fifty years later, in 2012, the Netherlands Institute for the Near East (NINO) initiated new investigations at Tell Shemshara, directed by Jesper Eidem. The project aimed to obtain archaeological evidence to supplement the rich historical information provided by the cuneiform archives.
Upon arrival, the team discovered that the site had suffered severe damage from decades of intermittent flooding. The high northern part of Main Hill, once excavated by the Danish expedition, had lost approximately four metres of elevation. Nevertheless, new excavations succeeded in re-exposing and documenting the Level V palace, and revealed a sequence of earlier Bronze Age levels (VIII–X) below it—including one isolated cuneiform tablet predating the Kuwari archive by some two centuries.
The project also conducted surveys across the Rania Plain, documenting the condition of archaeological sites within the flood-risk zone of Lake Dokan.
Re-excavating the palace of Kuwari (2014).
The Pisa Archaeological Project (2018–present)
Since 2018, research has continued under the University of Pisa, with the support of the Italian Ministry of Foreign Affairs and International Cooperation. Fieldwork has focused on further investigation of the early Bronze Age levels, which have yielded well-preserved architecture destroyed by fire, including large storage rooms with vessels still in situ.
Recent seasons have brought significant new discoveries. In 2023, excavations on the southern slope of Main Hill revealed the first documented evidence of Iron Age occupation at Shemshara, preliminarily dated to the Neo-Assyrian period (ca. 800–700 BCE). In 2024, a new excavation area exposed a well-preserved portion of the Level V palace courtyard, while work on the earlier Level VIII building was expanded to include five rooms of the palatial complex.
Room in Level VIII "palace" with large storage jars in situ, excavated autumn 2022.
View of Tell Shemshara (in the foreground) towards Lake Dokan and the Darband Pass.
Site Structure
Tell Shemshara is not a single mound, but rather a complex of natural hills that supported successive ancient occupations. The site is strategically positioned in an angle between the Lower Zab river to the east and the small Wadi Boskin to the west. Although the entire complex covers approximately 7 hectares, it is unlikely that all hills were simultaneously occupied during any single period.
Main Hill (SH 1)
The elongated Main Hill is the principal component of the site and preserves the densest archaeological deposits. It measures approximately 330 metres in length and 60 metres in width, with the northern end rising more than 10 metres above the rest. This was the focus of both the Danish (1957) and Iraqi (1958–59) excavations, and remains the primary area of investigation today.
The "Palace Terrace" on the lower, southern extension of the hill housed the Level V palace where the Kuwari archives were discovered. Recent research has shown that Bronze Age occupation on this hill was organised through a system of terraces cut into the natural slopes—a structural arrangement that unfortunately made the site particularly vulnerable to erosion from Lake Dokan.
Camp Hill (SH 2)
This large, oval hill lies northwest of Main Hill. The Danish expedition camped here in 1957, hence the name. In 1958, Iraqi archaeologists investigated the hill and uncovered remains from the Mitanni period (Late Bronze Age, mid-second millennium BCE), though details of this excavation remain unpublished.
Test trenches excavated by the NINO project in 2012–13 reached nearly 2 metres depth without exposing structural remains, but yielded Bronze Age and Neolithic sherds in what appeared to be natural soil. One trench near the high western part produced several reconstructable vessels and an early second millennium BCE seal impression in a trash deposit. The high western point of the hill hosts a small Islamic cemetery.
North Hill A (SH 3)
This elongated hill is located immediately north of Main Hill. An extensive Islamic cemetery covers its southern slope. There is no record of any excavation here during the 1950s campaigns.
Erosion scars on the eastern slope probably represent vanished Bronze Age terraces, of which only small traces remain. A test trench on the crest of the hill exposed sub-floor evidence for second millennium BCE occupation directly above virgin soil. A Neolithic terrace has also been identified at the northern end of this hill.
North Hill B (SH 4)
The oval hill immediately north of North Hill A is today heavily eroded and disturbed by numerous pits—reportedly robbed graves. Local informants mention finds of coins, suggesting an early first millennium AD date. The hill and its slopes are completely devoid of surface sherds and may not have had significant ancient occupation.
Gird Bardastee
Beyond the Wadi Boskin, to the west of the main cluster, lies a further hill known as Gird Bardastee. Iraqi archaeologists conducted excavations here in 1959. Just northwest of Bardastee, the NINO project examined a low, oval mound (ca. 200 × 70 m) with surface material indicating brief occupation during the Neo-Assyrian period (early first millennium BCE).
Stratigraphic Sequence
Excavations at Tell Shemshara have revealed a long but discontinuous occupation history, spanning from the Neolithic to the Ottoman period. The following sequence, presented from latest to earliest, is based primarily on evidence from Main Hill.
Traces of terracing and ephemeral structures on the western slope of Main Hill. Associated finds include clay pipes and iron objects.
Ottoman pipe-head.
Ottoman pipe-head.
Small village on Main Hill summit, possibly destroyed during Timur's campaigns in the late 14th century.
Identified in 2023 through test trenches on the southern slope of Main Hill. Excavations revealed three consecutive levels of stone foundations. Level ii, the best preserved, features walls of dry-stone masonry forming an elongated room, with a floor of clay plaster laid over a stone pavement. Ceramic evidence dates the occupation to the Neo-Assyrian period (8th–7th century BCE).
Assyrian ceramics from SP 2.
Drone image of SP 2 at the end of fieldwork (2024).
Attested only on Camp Hill (Iraqi excavations 1958). No evidence on Main Hill.
Fortified administrative complex with successive rebuilding phases. Discussed in detail in the following section.
Not apparent from either excavation or survey. A few Ninevite 5 sherds appear in later fills, but may have derived from Bardastee.
Earliest occupation at the site. Settlement terraces cut into the natural hill, investigated by the Danish expedition (1957) and the University of Reading (2012–13). New excavations by the Reading team identified foundational levels not reached by the Danish, extending the occupation sequence to c. 7300–6000 BCE. The earliest deposits, radiocarbon dated to c. 7330–7180 BCE, yielded well-preserved mineralised plant remains, mudbrick architecture, stone and pebble surfaces, and a rich assemblage of lithics, marble bracelet fragments and fauna including pig, sheep, birds and fish.
Cobbled surface dated to the LPPNB phase (Matthews et al. 2020: 184, fig. 10.8).
Neolithic sherd.
Neolithic sherd.
Neolithic labret found in later fills.
Middle Bronze Age Levels
The Middle Bronze Age occupation at Tell Shemshara spans approximately four centuries, from the late third millennium to the early 18th century BCE. Recent excavations by the Pisa Archaeological Project have revealed a sequence of six main levels (X–V) on Main Hill, documenting the site's transformation from a first fortified outpost to a regional administrative centre.
The following section describes each level in stratigraphic order, from the most recent (Level V) to the earliest (Level X).
The Rania Archaeological Project team excavating the Palaces Level V and Level VIII.
Level V represents the culmination of Bronze Age occupation at Shemshara: the palace of Kuwari, governor of ancient Shusharra. This fortified administrative complex, explored by Danish and Iraqi archaeologists in 1957–59 and re-investigated by recent projects, featured a central courtyard paved with baked bricks—laid in two successive phases—surrounded by multiple wings.
The east wing housed Kuwari's private quarters, including Room 2 where the famous archive was discovered. The north wing contained service areas, including a kitchen with storage installations. The palace was destroyed by fire, presumably during the rebellion led by Lidaja in 1780 BCE. This catastrophic event preserved approximately 250 cuneiform tablets that document a single dramatic year in the political history of the region.
Central Court (13) in the Level V palace, excavated in 2014.
Part of the western wing of the Palace, excavated in 2024.
Fragment of terracotta "tower", from fill in SW corner of Courtyard 13 (2014).
Under the Level V palace are two poorly preserved levels with several sub-phases, which do not yield coherent building plans within the present exposures. Both levels clearly belong to the Middle Bronze Age.
Sandwiched between the oldest phase of Level VII and the ruins of the partly burnt Level VIII it seems the site was used as a cemetery and graves have turned up in several areas of the excavation
Grave on top of Room 19
.
Grave in Room 12a.
Grave without offerings in fill above Level VIII in north part of Room 4.
Updated plan of Level VIII Palace.
Level VIII represents the best-preserved early occupation at Shemshara. A large building, probably palatial in function, was constructed using a distinctive regional technique: pisé (rammed earth) interspersed with courses of mud bricks, set on shallow stone footings. Several rooms have been excavated, all showing evidence of heavy burning and two successive phases of use.
Storage facilities were a prominent feature: large ceramic jars were found embedded in floors, some still in situ. One isolated cuneiform tablet recovered from this level—predating the Kuwari archive by some two centuries—demonstrate that Shemshara already played an important administrative role in the Rania Plain during this early period. The tablets mention individuals from Kakmum and Urbilum, two major polities of the region.
Room 1a.
Room 16-19 (excavated 2024).
Room 3 (excavated 2013).
Room 1a during excavation (2022).
Room 12a (2021 excavation).
Cuneiform tablet from Room 1a (2012 sounding).
Overview of Level IX "River Gate" (2013).
On the eroded north and east slopes of Main Hill are remains of a fortification system pre-dating Level VIII. On the North slope a min. 4 m wide wall of mud bricks and with rooms behind it have been exposed close to edge of the hill. On the east slope a large terrace cut into the natural hill was supported by massive retaining walls, and on the lower slope are remains of another fortification wall of red bricks with what seems one side of a gate with a small tower. Ceramics from Level IX can be dated to the outgoing 3rd Mill - very early 2nd Mill. BCE.
Level IX brick core in the Southern retaining wall (2013).
Fortification wall on north slope (2013).
Eroded Level IX rooms behind the fortification wall on north slope (2024).
Level X remains under Level IX north tower of "River Gate" (2013).
The earliest Bronze Age occupation at Shemshara remains elusive, defined only by a few features sealed beneath Level IX. The ceramics found in these early contexts suggest a date in the late Akkadian period.
Eroded remains of Level X behind Level IX enceinte at foot of north slope of Main Hill.
Level X stone foundation below the later Room 8 of Level VIII.
Level X wall under later Room 18 of Level VIII.
Palace & Archive of Kuwari
The cuneiform archives discovered at Tell Shemshara remain unique finds in this corner of the Ancient Near East. Approximately 250 clay tablets, inscribed in Akkadian, document a single dramatic year (ca. 1780 BCE) when the site—ancient Shusharra—became the centre of regional and international political developments. The ancient name remarkably survived into modern times as "Shemshara".
The tablets belonged to Kuwari, governor of Shusharra, and reveal how he navigated a dangerous political landscape caught between rival powers: the expanding empire of Shamshi-Adad in the west and the mountain kingdoms of Itabalhum and Gutium in the east.
Updated plan of the Level V palace.
The palace of Kuwari was a fortified administrative complex organised around a central courtyard paved with baked bricks. The east wing housed Kuwari's private quarters, including Room 2—the so-called "Tablet Room"—where the Danish expedition discovered the main archive in 1957. The space to the south of Room 2 was a small courtyard, from which a gateway led west into the large central court.
The north wing contained service areas, including storage facilities with large ceramic jars embedded in the floors. A second group of tablets was found by Iraqi archaeologists in 1958 in Rooms 27 and 34, immediately south of the central court. These rooms were accessible from Room 30, which connected to the main courtyard. The entire building was destroyed by fire, presumably during the rebellion of 1780 BCE, and the conflagration paradoxically ensured the preservation of the clay tablets.
Fragments of letter envelope with impression of the seal of Pishenden, king of Itabalhum; found with archive of Kuwari at Shemshara, 1957 (Eidem & Læssøe 2001: 159 no. 1).
In mid-July 1957, just days before the end of the season, the Danish Dokan Expedition uncovered an archive of approximately 150 cuneiform tablets in Room 2 of the Level V palace. The room measured 3.55 × 4.70 m and had a floor of reddish baked bricks (39 × 39 cm). Besides the tablets, excavators found a ceramic jar stand, a fine ware beaker, and the torso of an animal figurine. Jørgen Læssøe, an expert in ancient languages, quickly recognised that several tablets were letters sent by the famous Assyrian king Shamshi-Adad to a man named Kuwari.
Political unrest prevented the Danish team from returning in 1958, but Iraqi archaeologists continued the excavation. They exposed much of the palace and discovered a second archive of about 100 tablets in Rooms 27 and 34. These were mainly administrative lists concerning agricultural products, providing valuable information about the local economy and the rural hinterland that Shemshara controlled.
The main reception room in the palace (Room 35, Iraqi excavation 1958).
Terracotta stand found in Room 35 (Iraqi excavation 1958.
Room 32 (Iraqi excavation 1958).
Examples of administrative tablets in the archive of Kuwari.
Kuwari—a Hurrian name—was placed at Shemshara as governor by the kingdom of Itabalhum, a polity in the mountains of western Iran. At least 66 individual letters in the archive were addressed directly to him, confirming his role as the owner and recipient of the correspondence. He exchanged envoys and letters with kings and officials in the Turukkean realm in the Zagros and, later, with the mighty Shamshi-Adad and his generals in the west.
The administrative texts reveal that Shemshara controlled a fairly large rural hinterland known as the "land of Utûm", which likely incorporated the southern sector of the Rania Plain and perhaps a western portion of the Pishdar Plain to the east. Documents record the distribution of metal items, weapons, garments, and shoes, as well as lists of agricultural products—evidence of a functioning palatial economy.
Letter from Hulukkatil (an eastern associate) to Kuwari (Eidem/Læssøe No. 50).
Letter from Kurashanum (official of Shamshi-Adad) to Kuwari (Eidem/Læssøe No. 31).
Drawing of the obverse of tablet no. 31 (Eidem & Læssøe 2001: pl. 31).
Geopolitical map of Eastern Mesopotamia at the time of Kuwari.
The tablets document a period of intense political turmoil. Itabalhum, the kingdom that had appointed Kuwari, was at war with another mountain kingdom, Gutium. At the same time, Shamshi-Adad of Assyria was conducting a major military campaign in the region around modern Erbil, rapidly approaching the Rania Plain. A friend wrote to Kuwari with alarming news: Shamshi-Adad had conquered Arraphum (Kirkuk), was moving towards Qabra (Erbil), and had sent his son with 60,000 troops to besiege Nurrugum (Mosul). "Hopefully Shamshi-Adad will not conquer the whole country," the letter concluded, "and we shall not have to worry."
But then came worse news: the army of Gutium had raided the core of Itabalhum. Shemshara was now in a perilous position—who would reach it first? The local population chose to seek protection from Shamshi-Adad against Gutium, and Kuwari followed them. The site transformed from the western outpost of a mountain kingdom into the eastern outpost of Shamshi-Adad's empire. Instead of receiving letters from friends in the east, Kuwari now received orders from Shamshi-Adad and his generals.
The arrangement did not last. When news arrived that Shamshi-Adad planned to make a coalition with the king of Gutium—offering his daughter in marriage and "the country of Shemshara" as her dowry—the people rose in rebellion against both Shamshi-Adad and Kuwari. A letter from Mari reports that "the land of Shemshara is troubled, and we cannot hold it." The palace was destroyed, presumably in this uprising led by a man named Lidaja in 1780 BCE. Whether Kuwari escaped remains unknown. In the aftermath, some 4,000 inhabitants of the land of Utûm were deported to Ekallatum, the Assyrian heartland.
Following the division of finds after the 1957 excavations, approximately half of the tablets from that season were allotted to the Danish expedition and are now housed in the National Museum of Denmark in Copenhagen. The remaining tablets from the 1957 campaign, together with all of the approximately 100 tablets and fragments found during the 1958 excavations, were kept in the Iraq Museum in Baghdad. Some of these have since been transferred to the Sulaymania Museum, where at least four are currently on display: three letters and one administrative text.
The complete archive has been published in two volumes: the administrative texts in J. Eidem, The Shemshara Archives 2: The Administrative Texts (Copenhagen, 1992), and the letters in J. Eidem and J. Læssøe, The Shemshara Archives 1: The Letters (Copenhagen, 2001). Together, these documents provide an unparalleled window into the political history of this remote corner of the Ancient Near East during a single dramatic year nearly four thousand years ago.